PVCXOGONIDA. 



33 



based on the fact, whether the foremost limbs, as well the embryonal limbs as the imaginal limbs, 

 continue the typical development, or their development is checked on an earlier or later stage, whether 

 they go through a slighter or more marked retrograde development, or how long they are kept upon 

 the whole. How small a systematic importance these foremost limbs have, may be seen, among other 

 things, by the fact that a pair of these limbs may be kept in one sex, while it is thrown off in the 

 other, and thus the reason of the keeping seems only to be purely biological; moreover the keeping 

 or the throwing off takes place without producing or corresponding to any other difference in struc- 

 ture between the two sexes. To make such organs the base of the division into orders, seems to me 

 very unfortunate, and it is only for want of better characteristics that I have used these limbs as 

 family distinctions — in reality they are only of value as generic distinctions. Therefore I do not 

 think it necessary to enter upon a detailed valuation of the orders that have been put up. 



Among the different systems those of Wilson, Syn. Pycnog. New. Engl. 1878, and Hoek, 

 Report Pycnog., 1881, and Nouv. etud. Pycnog 1881, seem to me to be the best. The system of 

 Wilson must be said to be well worked out, but on the other hand it is rather artificial, too much 

 stress being laid upon characteristics taken from the auxiliary claws and the number of joints in the 

 palps and the ovigerous legs. The system of Hoek in Report Pycnog. is merely a grouping of the 

 genera without any real arranging of these genera inside the families. In his system in Nouv. etud. 

 Pycnog. p. 106 the number of families has been reduced to four, and these four correspond to my 

 system, as well with regard to the appellations as, chiefly, to the characterization, and the genera 

 contained in each of them; it is however to be noted that the genus Pallaiofisis has by Hoek been 

 referred to PJwxicliilida , but by me to Nymfihonidtz. The genealogical table given by Hoek with 

 its extremely problematic primitive form Archipycnogonum, I am not able to appreciate rightly. 



The system I have used, is more particularly intended for the Pycnogonida brought home by 

 the Ingolf expedition with the object of its also being able to comprise the new species. 



I. Fam. Nymphonidae. 



$m. 



Corpus manifeste in segmeuta partitum. 



Rostrum cyliudricum, inflexibile, libratum vel nutans. 



Chelifori expleti, chela instructi. 



Palpi vel expleti, vel imminuti, vel deficientes. 



Pedes oviferi in utroque sexu. 



The trunk distinctly segmented. 



The proboscis cylindrical, inflexible, horizontal, or directed obliquely downward. 



The chelifori well developed (with chela). 



The palps well developed, or rudimentary, or wanting. 



Ovigerous legs present in both sexes. 



As it will appear from the family-diagnosis, this family comprises the species belonging as 

 well to Nymphon as to Pallene, that is to say, the families Nympliouida and Pallaiidw of Sars. An 



The Ingolf-Expedition. III. i. 5 



